Molten salt reactors would be quite good at load following, because the reaction rate slows down so much as the fuel temperature increases. If you draw less power, you apply less cooling, so the fuel heats and the power output goes down. For example, the IMSR, likely the first MSR to be commercialized, is naturally a load-following reactor [2].
This doesn't mean it wouldn't make economic sense to, say, run nuclear for baseload, add solar for extra daytime demand, and use storage to smooth out the remaining demand mismatch. But that's a matter of economic optimization, not a technical limitation of nuclear.
[1] http://ansnuclearcafe.org/2013/02/14/responding-to-system-de...
In contrast, we know with near certitude that investment in basic infrastructure, such as a capable and robust power grid, will promote a plurality of solutions and pay for itself over time.
Yup. Larger power markets benefits every kind of power: wind, solar, and nuclear. There is a lot of resistance to this though. Every state would prefer wind sites build in their own state, rather than buying power from the windier state farther away.
If you had a truly national sized grid then the power distribution between even just wind and solar is remarkably stable.
Either way, we need a larger and more capable grid, so that production and demand can be balanced over larger areas.