Well, it beats making them into the world's most complicated air planes as with the space shuttle. SpaceX has proven that they can do vertical landings of the first stage intact onto both land and a seagoing barge; after a trip out of the atmosphere and to about 1/5 of orbital velocity but not into orbit.
It's great to see the myths of non-reusable first stage technology being dispelled.
That turns out to be far less expensive and complicated than a space plane. It does turn out we need a lifting body for much larger vehicles. It still doesn't have to be a plane, though.
I think you have to separate the concept from the implementation of a space plane design. SpaceX's launcher effort is less expensive and complicated than the space planes implemented so far. However both Buran and the Space Shuttle had the orbiter on the side of the stack where their heat shield was exposed to debris from the launch vehicle. So much so it was considered to be 'In-Family' because it would occur every launch to some degree.
The Space Shuttle also had all of the complexity related to the main engines which had no business being in orbit. Having the mass of those engines also complicates the infrastructure of landing. Considering also many of the political issues that interfered with the implementation of the Shuttle it wasn't the best design for a space plane.
They are useful though. The contrast is having the space plane at the top of the launcher stack, which removes the 'in-family' debris, main engine mass and, complexity issues for operating a space plane, as the X-37 demonstrates. The X-37 is covered by a shroud, however having some control surfaces at the top of a launch stack for a larger vehicle could be useful considering launchers have control surfaces near the base to balance it. A less massive orbiter means more for vehicle and payload reaching orbit and returning.
Musk has taken the pork barrelling out of launcher design and shown that it is possible to recover launcher assets. That knowledge maybe the foundation of opening up commercial space ventures paving the way for more commercial launcher platforms to become available.
The good thing about commercialization could mean someone implementing a space plane with a better design or more specific role. With those things in mind maybe we haven't seen the most optimal design for a space plane yet.