"The steam cycle is a bird in the hand. In terms of locations, most new generation will have to be concentrated in existing sites, which already have cooling water."
Let me throw this out.

Many coal plants are being retired due to EPA regulations. It is not cost effective to retrofit these old boilers to have reduced mercury, SO2 and particulates. In the next few years, there will be older turbine generators that are surplus. These will probably be just a bit superheated since they are old.
Older steam plants have been retrofitted to become combined cycle units by adding a gas turbine and using the waste heat in the exhaust gas to boil water and replace the coal boiler. Why not retrofit an old steam plant by replacing the boiler with a LFTR?
Since the fuel is cheap, you don't need the extra efficiency of a super-critical unit. Cheap BTUs will be produced with no emissions. The utility will be happy to buy this power. Many utilities will be very dependent on natural gas in the near future. Having all their eggs in one basket (natural gas) does not give a comfort factor to these folks.
A lot of money could be saved in a demonstration project by retrofitting and old site and proving that a LFTR can be a reliable replacement for the steam boiler. The turbine, generator, balance of plant equipment and substation will be there. This is a savings of millions of dollars. It may even be possible to use the support steel for the boiler to house the LFTR.
Would this be good recycling or an additional problem for the new LFTR project?