Costs were ballooning even before TMI.
> The other part is that real coal prices fell in the 1980s instead of rising.
More importantly, 1979 saw the passage of PURPA, which began to open the power market to non-utility providers. There was enormous untapped potential for cogeneration (and, as it turned out, cogeneration-in-name-only) that produced a slug of new output, mostly gas fired, into the grids just after what had been inexorable 7%/year increase in electricity demand in the US suddenly moderated.
In this environment, it was very difficult to make the case for new nuclear power plants.
> I wonder if those over-optimistic solar cost predictions you saw in 1975
In what sense were they over-optimistic? PV has experienced a remarkably relentless cost decline along an experience curve of about 20% decline in cost with each doubling of cumulative production.