You raise some good points. First, the proposed heavy lifter is being designed out of mostly Shuttle components. I know NASA has a poor track record managing large projects, but how much R&D is needed for a 5-segment SRB (already tested) and simplified Shuttle Main engines (developed almost 30 years ago)? There really isn't much new technology in the design and that's the whole point.

There are some areas that would get expensive. For example, you'd need to either perform extensive (and expensive) modifications on existing launch facilities or build a new one. You'd need to develop the flight software (remember the lesson of the Ariane 5's first flight) and some other things. Still, developing a heavy lifter from Shuttle components shouldn't cost too much.

As for multiple segments in a lunar or interplanetary vehicle, it presents challenges but also an opportunity. The challenges are that you'd need to develop modules for orbital assembly. As the ISS experience has shown, this isn't easy. If you launch the modules on unmanned vehicles (like the Russians did with the Mir modules), then each module would need it's own TT&C, power, propulsion and guidance systems to be able to fly itself to the assembly point. All of these systems make each module more expensive and complicated and, once the vehicle is assembled, represents dead mass. I wonder how the extra R&D plus acquisition cost for all of those modules of flying a couple missions would compare to the cost of developing the heavy lifter.

As for the propellant, you'd certainly launch that last. However, you'd probably want to use less energetic hypergolic propellants instead of cyrogenics due to the boil-off issue you discussed. With cyrogenics, you'd need to develop tanks that have reduced boil-off similar to what they use on the Shuttle for the fuel cells. However, those dewars are much heavier and more expensive than a simple propellant tank. Without this capability, you'd need a launch surge ability to get several loads of propellant up there and docked within a matter of days or at most a couple weeks. That surge ability means you'd probably need more launch pads, crews, flight crews, etc. to pull it all together. In short, you'd have to build a lot of expensive infrastructure to support a surge that might only happen once every couple years. In addition, you'd have many more "opportunities to excel" (screw up) due to the heavy launch tempo. "Go fever" often leads to expensive and sometimes tragic mistakes.

The one positive aspect of multisegment vehicles is the potential for replacing modules on a reusable design. However, nothing I've seen indicates that they plan on reusing anything other than the CEV capsule itself. IMO, that's a damned shame.