The last Falcon launch brought JCSAT-16 to a supersynchronous orbit, very definitely not LEO, with the apogee at 36183 km and the apogee 151 km, and about 20 degrees inclination off of equatorial. The apogee was a bit higher than geostationary. The remaining load for on-board propulsion is to change the inclination (which is most economical to do with a burn at apogee) and to circularize the orbit (raise the perigee).
By giving the satellite a kick to high orbit, the Falcon 9 saves fuel on the satellite that will be used to maintain the orbit longer than would otherwise be possible.
I have always been highly skeptical of are their launch costs. I simply don't believe them.
If you are concerned about the "shear violence", I suggest you go to 1 Rocket Rd, Hawthorne, CA, cross-street is Crenshaw. Stand in front of the building. SpaceX has left a rocket right on the front lawn for you to look at, a first stage that returned from lifting the Dragon capsule to ISS. It got to 1/5 orbital velocity (the second stage does the rest), burned its rockets for about 2.5 minutes, was in the air for less than 10 minutes overall.
Regarding the economics, I think the main point is that there was not an incentive to lower cost until now. The USA had a single-source contract and the two former competitors formed a joint venture so that there would be no competition. Also, there was more subcontracting: for example most companies didn't make their own avionics and these came with tremendous markups, space-qualified fasteners were quoted at $10/screw in the '90's and are probably more now.
So, a vendor who actually tries to reduce prices can probably reduce them a great deal, simply because nobody else has tried very hard before. There would be a lot of low-hanging fruit.
I agree that Elon is way too self-indulgent. Forget about the simulation remark, hyperloop is either cynical in nature (meant to divert funds from real trains) or wildly underestimating the costs and safety issues.
However, I think you're wrong about the space junk issue. One of the problems right now is the lack of any way to economically de-orbit legacy space hardware in high orbits. You don't get that without economical access to space.
The reason that you spent that much money building the cargo has comparatively little to do with the cost of the launch and everything to do with the fact that you really don't get multiple chances to get it right
I think you need to go back to your initial assumption, which might not be true any longer. With lower $/kg to your selected orbit, replacing a satellite is economically possible and building a satellite with a much shorter projected lifetime is probably optimal because the alternative is for the operator to be stuck with 20-year-old technology in orbit (given 15-year design lifetimes and a 5-year design-to-launch cycle, which might be optimistic).
If you really run that sort of company, you need to be looking about what could happen if your assumptions are wrong. And not advertising the fact that you aren't doing that.
What is the limiting factor? Buildup of CO2?
People need a certain amount of oxygen for their metabolism, you need to carry that much. CO2 effects the blood pH: too little and the body is too alkaline, too much and it's too acidic. So, you need to maintain a precise amount of CO2 and remove the rest. The scrubbers in the space shuttle were able to regenerate the CO2-absorbent material after use, so there was use of power but material wasn't consumed.
Beyond this, you need to control temperature and humidity. The other requirements than atmosphere for crew survival are that you water, feed and shelter the crew, maintain orientation, and maintain a G-force envelope that doesn't injure the crew.
This universe shipped by weight, not by volume. Some expansion of the contents may have occurred during shipment.