Some new technologies are great on a bang for the buck basis. Some are hideously expensive for marginal returns.
I say this will all deference to the problems NASA faces: One of their greatest failures is going off the deep end of diminishing returns.
An interesting tidbit: Even in the 60s, without all the advances in materials and structures, SSTO-like mass fractions were very close even in non-SSTO applications. Yet the conventional wisdom remains that SSTO is technically unfeasible.
One of the problems with the space industry is that it really is hard to do on the cheap. For small payloads, $/lb is ruinous. For really large payloads, $/lb improves dramatically, but not enough to offset the ruinous mission cost.