Future mortality seems not to be considered but I am afraid this is important. Numbers will be difficult to obtain, you know, clairvoyance is not easy.
But at least we know that we need to maintain nuclear waste for a long time. Or we create deep subterranean disposal sites, and this will be dangerous work.
Coal is probably even a lot worse because of climate change effects.
Renewables seem to fare better for future mortality.
I do agree that the overall risk includes more factors than just past direct/indirect deaths, you are right. However, I would argue that simply furthers the cause for nuclear.
Firstly, renewables are great and we should of course be investing heavily. However, until we have the what is still non-existing storage tecgnology, we need a base load of either nuclear or fossil fuels. Insisting on a nuclear-free energy supply right now unfortunately means insisting on fossil fuels as a base load for the foreseeable future.
Given the greatest risk to humanity's survival is climate change, failing to secure a low-carbon energy supply is a risk that far outweighs any of the risk of using nuclear alongside renewables.
We can already see the results of foregoing nuclear - energy in now nuclear-free Germany is seven times as carbon-intensive as nuclear-heavy France[1].
The problem is people like you redefining what risk means, then saying the people who define it properly don't understand it. The problem is people with small minds or an agenda on their small mind, redefining what is important to others, and calling them stupid. By your false definition, the risk of me getting hit by a car when crossing a street is less than the risk of me getting killed by the smog cars make. While quantitatively true for a statistical population, it is qualitatively false. And people's opinions are qualitative - they're based on a definable reason, not just explainable symptoms. Guess what: black people are 15% of the population but commit 50% of the violent crime, according to FBI statistics. Quantitatively. Does not mean that statistic applies to a single random person on the street and he can be arrested because he's likely to commit a crime. That's your logic applied to risk of getting stabbed.
Here's the risk as society that doesn't pretend to be purposefully dense defines it: when something goes wrong, how bad it is. Take all the coal plant disasters, and compare them to all the nuclear disasters. Now let's take a future disaster, the reason for which you don't know and cannot account for with "new design" which will be called "old and faulty design" in a few decades. During that disaster, looking at past disasters, do you want that destroyed power plant to be coal or nuclear?
Coal has no risk. Coal has a well defined, predictable, and understood small and slow detriment. You define that as risk. The world defines that as the opposite of risk.
Lived in Kiev for a year an a half, took a bus trip to Chernobyl.. Guess what they got in that huge area where no one lives (actually there is a crazy old lady who lives there, still in her house). Did you guess it? Yeah, fresh tree stumps.
That radioactive wood is cut down by shady companies for free, and shipped to europe and other parts of Ukraine. Out of it you get houses and furniture. Did you account for risk of sitting on a radioactive couch in a radioactive house when you claimed you fully understood "risk?" Ah, that's because to understand the risk of something you first need to understand what the English word means.
The problem, as you point out, is that people are generally very bad at understanding risk.