Barely.

Even on Earth we have cooling towers.

At the very least, we'll have a better idea what our real needs are. Right now we're like Christopher Columbus trying to design the NY subway system.

The space shuttle was a 100+ ton space-plane launched on a Saturn V class launcher. Built without any real precursors, from 1970s technology. Every aspect of it pushed the technology beyond the state-of-the-art. No part of it was built to reduce operating costs. The original proposal sold to Nixon may have been to develop a low cost "space truck", but that was never part of the actual program development goals.

OTOH, Falcon 9 was intended solely to be cheap. And is already the cheapest launcher on the market as an expendable. Even partial reusability (first stage) is expected to lower launch costs significantly. Musk claims that launch operations costs are a small part of his launch costs, and even that will probably drop once his team controls their own site and range.

This gets back to moving away from the "standing army" model of spaceflight operations.

Very large radiators.

And the radiators must be protected behind a heavy shield, because the radiation degrades them too quickly. Most designs have the reactor, then a heavy shield, a long truss and then the rest of the ship. Running down the length of the truss, carefully shaped to remain in the shadow of the shield, you have huge radiators to dump the heat from the reactor. The truss, the radiators and the shield are all additional mass required for a nuclear propulsion on top of the reactor mass. Solar arrays require radiators too, but only a fraction of the size, see the ISS.

In a way they did. They sent expendable seamen and some Italian loudmouth on a few small off-the-shelf commercial ships. Only when the destination was proven did they risk their more valuable people on more valuable ships.

Ummm...

...yeah...

That's the problem. You can't experiment where people live, or where there are things you care about.

that only puts a few hundred to at most a thousand-plus kg of payload on the moon per flight.

You're doing it wrong. With reusable launchers you want to split your payloads into cargo and fuel, launched separately. Most products have a U or J shaped reliability curve. Higher failure rate at the beginning, picking up manufacturing defects, then lower failure rate until end-of-life effects start to accumulate. Because fuel is cheap, you put that on the brand new, untested reusable launchers, and on the end-of-life launchers and fly them until they die. Because cargo is generally not cheap, you use the in-between launchers, at their peak reliability.

You put your cargo and TLI-engines into LEO on the most reliable launchers using the payload full capacity of your launcher, then you add fuel tanks and associated plumbing on the less reliable launchers, finally you launch bulk propellant on the lowest reliability launchers. Then you launch these large payloads from LEO. You lose some Oberth efficiencies, but you gain in using your launcher's maximum lifespan, and being able to launch larger individual payloads (in the 10-12 tonne range for F9, 50 tonne range for FH, and 250 tonne range for the Raptor-based MCT.)

No form of currently-achievable propulsion yields a higher Isp than a fission fragment rocket

We're so far from FFR, we might as well talk about fusion drives, or Harold White's warp drive.

My previous comments apply to NEP vs SEP. SEP has better power/mass ratios until you are somewhere near Jupiter, and realistically probably somewhere past Jupiter.

I've never understood this argument. By definition, you are saying that everything that is available to these private players was already available to NASA and its primary contractors.

So why is that a justification for minimising the achievements of the new guys? "Yeah, we could have done that. We didn't, but we totally could have." No, no you couldn't. You had decades to do it, your whole job was to do it, you had billions in funding to do it, your primary contractors are orders of magnitude larger than the new players, and yet you didn't do it.

Worse. Even after the new guys did it, you still aren't doing it.

For example, Boeing is getting a multi-billion dollar USAF contract to develop a new large methane engine. They got the contract because of their "decades of experience". Aaaand... they are sub-contracting the actual work to Blue Origin, a dot-com billionaire's tiny little hobby-project which has spent the last decade actually building new rocket engines. But hey, it's the new guys who are leaching off the old players. Sure.

Most things can be broken down into sub-100 tonne parts. The problem is not the launch, at least once we stop throwing away our rockets, it's the cost of on-orbit operations. That requires a change in how we work. Ending the standing armies on the ground to support every spanner-turn in space.

Trade studies have suggested that out to the main asteroid belt, aerospace grade solar panels have a higher power/mass ratio than nuclear systems. Only out near Jupiter does the equation shift (but even that is only counting the direct reactor mass. The added mass of shielding, trusses for distance, etc, is usually not included.) And every year, the cross-over distance shifts further out.

The exception is where sunlight is unavailable — Lunar night, Mars winter — where the length of darkness exceeds likely storage capacity. However, the most likely location for early development on the moon is the poles, where there are Peaks of Eternal Light. (How can you not capitalise that?) OTOH, for Mars, you are probably going to avoid the poles due to the severity of those winters, staying within 30 deg of the equator, avoiding that problem too.

So it'll be a fair while before we need nukes, better to focus available funding on something else.

Solid metal will pass through virtually unchanged. Shaped for minimum-drag in hypersonic entry (unlike capsules, which are shaped for braking), it will, by coincidence, be virtually undetectable by radar or optical until it hits the atmosphere. (Unless they are watching activity at the launch site, of course. Which they would be, if they aren't stupid.)

No-one saw the 10,000 tonne, 20m diameter Chelyabinsk meteor, which produced a half-megaton detonation. It was a stony-type meteor, hence it didn't penetrate below 25km. A solid metal manufactured impactor could easily reach the ground; but the optimum shape for a city killer would detonate in the air above the target, but below 5km.

To reach the moon with even a small nuclear warhead, you would need a full multi-stage rocket launch, not a little ICBM squirted out of a sub or silo. You couldn't hide such a launch. And it would be trackable for the entire 3 day transit to the moon. The lunar loony would just need to launch a load of gravel back along the same trajectory. No more warhead. He then throws some impactors at the launching nation's small number of suitable launch sites. No more counter-attacks. He can also hit a few LEO satellites to start a Kessler cascade, preventing any launches from Earth for several decades. His own impactors (being purely bulk-kinetic) aren't affected as they pass through.

The moon has "air superiority" over Earth due to orbital mechanics. It takes a lot more energy to bomb the moon than it does to bomb Earth. Therefore attacks from Earth must be much larger, and require substantially more infrastructure to launch, which is vulnerable to attack from moon-thrown spears-of-death. (The moon's disadvantage is that it takes a lot of infrastructure to get to that point.)

International Traffic In Arms Regulations

I think you replied to the wrong comment.

Also what does "Mars mission" mean in the context of this question?

Permanent base on the surface? Flags and footprints? A mere fly-by? Landing on Phobos and restarting the reactor?