It hasn't priced itself out of the market - Governments insisting that it is many orders of magnitude safer [1] than any other form of power generation have unsurprisingly made it very expensive and over-engineered.
Nuclear quite simply needs to be given its fair share of blood to be competitive - to get wind power down to the same level of safety we'd have to build robots to climb the turbines to maintain them.
The argument that governments aren't letting reactors melt down enough is a nonstarter. If nuclear proponents have to resort to that, you might as well give up.
I suggest instead pushing for designs that can achieve acceptable levels of safety at lower cost. In particular, molten salt designs, such as the one from Moltex, that allow the size of the containment building to be shrunk by a factor of five from that of an LWR (for a given power output). Moltex's design does not dissolve the fuel in the primary coolant and has other attractive features vs. other MSR designs.
Such designs are not on the market today, and will take time and money to bring to market. I think R&D on them is a good idea, and supporting that is a proper role for governments, just as it was for renewables, but they are still something of a longshot, given where the competition is likely to be in 20 years.
It isn't just the reactor that is over-specced - anything to do with nuclear is held to a much higher standard than anything else, relative to the actual risks.
Thorium molten salt reactors have been bandied around for a while, especially in India due to necessity, but I fully expect governments to still make unrealistic demands of them, and therefore drive them up to unrealistic prices.
Stuff like this depresses me - I don't see how politics can handle grown-up conversations about trade-offs like this when people will just vote in the cake-and-eat-it parties.