We need to examine our design philosophy, because using the highest performance possible is expensive, difficult, and the advantages the highest possible performance provide can be outweighed by simplicity and reliability. Keeping a rocket on the pad fueled and ready to go as soon as the weather breaks is a lot cheaper with LOX and RP1. Neither is difficult to handle, which makes routine procedures a lot easier, and using more engines provides adequate thrust. When we were fighting every pound, trying to maximize the potential payload for a extended mission, it made sense to use LH2. But we just need to get the stuff to Low Earth Orbit, and then we can move it with an Orbital Transfer Vehicle, or couple it to the drive module, or whatever, but LEO is all we have go.
Build in a launch program which will absorb development costs while lifting fuel, living quarters, work shops, a hanger, and a few OTVs, and request delivery dates which promote assembly line construction. Buying the first 10 or 12 copies will make the development more affordable for the company building the rocket, and establish the capabilities of the launch vehicle for the world market. Once we have a way of putting most of a space station in orbit in one launch, I believe that we will see a lot of interest in building space stations.
Of course, we need to couple this heavy lift launch capability with the ability to put lots of people into orbit, so that there will be people to do research, people to help those people, people to keep those people alive, and people to keep the whole thing working. We need a bus, a cheap and easy way to get people into space, without going through the hassles of man-rating a rocket to take off straight up.
See the thread 'A cheap and easy way into space' in this forum for more info.