I would just point out that we really have no idea what we will have with a fusion power plant. We think all those things will be true, but we won't know for sure until we build one.
Only 20 years away so it won't be long.
Actually, the parameters of the supporting design for a fusion reactor are pretty well scoped out, just as they were with the first fission reactors. Those early designs, without a tremendous amount of revision, were the basis of the Shippingport reactor, the first commercial production reactor.
It's extremely doubtful that, after nearly 60 years of research and experimentation, they're going to go "OMG! OF COURSE! IT NEEDS TO BE 72 TIMES MORE COMPLEX, 150 TIMES BIGGER, AND 7,000 TIMES MORE EXPENSIVE!"
Yes, iterative design will continue to move costs upwards as we get closer to solving the sustained fusion reaction problem but, there are some estimates that a fusion reactor could last (with maintenance and replacement parts, of course) for 100 years or more.