Yes. The problem with nuclear is basically twofold:

first, the approval and regulatory process is deliberately cumbersome and in obvious need of reform. Treating every plant as a one-off design rather than standardizing has enormously inflated costs. And generally much higher scrutiny requirements for new designs have strangled the ability to roll out better/safer designs. It's very similar to what happens in the FAA with aircraft/engine designs, we can design much better engines/reactors than we could in 1960, but the new ones require an onerous approval process while the old ones got grandfathered approval. So we only build the old/worse ones!

If you don't want any more nuclear constructed, set policy to that effect, don't use the approval process to artificially inject costs to make it unfavorable. And the disposal situation just needs to happen, period. The waste has to go somewhere, we can't just have it sitting around forever. Even if we never ran another nuclear plant ever, the existing waste still has to go somewhere, and that process was dragged to a halt for political reasons too. The Yucca Mountain repository needs to be moved forward again.

Second, we need to stop letting coal externalize its costs. Tax carbon emissions heavily, require secure disposal of radioactive coal ash rather than letting it sit around in storage ponds that eventually spill and destroy miles and miles of land, etc.

Unfortunately, in both cases, the fossil fuel industry has its finger on the pulse of washington and isn't going to allow a trillion-dollar industry to be torn down without a massive fight. Existing stakeholders are just too entrenched for it to ever be successful and construction costs/disposal/externalities are just the place where that iceberg breaks the surface.

The "once you build the expertise, costs come down for the Nth plant" is also probably true, but there's also other things going on here with the process as a whole.

The article hints at the core driver of costs: time

Decrease time-to-completion, and you decrease a whole host of costs. And the regulations driving that time explosion are within Congress's ability to modify.

(1) Grandfather any currently builing plan into regulations as of the year of project initiation, regardless of how they change after the fact

(2) Waive environmental impact studies for nuclear (or other no-carbon) plants. Don't let the perfect be the enemy of the good

The real problem is that nuclear has no constituency. Environmentalists hate it because they can't do math and aren't pragmatic (let's be honest); wealthy folks hate it because they don't care how much power costs and NIMBY.

So who exists to fight for it?

The only political hope I see in the US would be a single-issue group comprised of practical environmentalists + industrialists. The former should like it because it saves the planet. The latter should like it because it allows decarbonization without inflating energy costs.

You nailed it about aerospace: the manufacturer I work for makes aircraft that were certified in the 1950's. The visitor entrance to the production facility has the type certificates framed near the entrance. We routinely design "new" aircraft that are basically lipstick on the old grandfathered in designs from decades ago. Whatever it takes to not have to recertify.

Doesn't seem to have helped any at Vogtle.

One of the victims of coal was Pat McCrory in the 2016 NC governors race. Signing “H.B. 2,” the stupid transgender-bathroom law, is often cited as the main reason for his loss, but I believe the coal-ash spill[1] and subsequent coverup scandal were decisive in his losing support in rural areas.

I certainly agree: the true costs of coal should be better believed!

https://en.m.wikipedia.org/wiki/Pat_McCrory#Duke_Energy