I believe a better perspective is would be the least harm perspective. What causes the least harm? Living with 1800s technology and no electicity, coal, nuclear, green, etc along with the various tradeoffs and our current level of technology.
Just because nuclear has the potential for harm, once that harm is computed and compared to other possibilities, we have an answer.
Big blowups are bad, but if they are rare enough, they are not really a problem.
In theory, I agree. But I doubt this risk consideration can feasibly be performed and lead to action in practize. How can we measure the damage resulting from fossil fuels, for example? Although by far most scientists on that field agree that is the case, there is not even a majority accepting that CO2 emissions lead to climate change. And even acknowledging that makes it impossible to really calculate the costs resulting from climate change (may be relatively benign, or very high, or bad for others yet not so bad for us -- depending on the models you use for the calculation). And how can we really trace back pollution related illnesses, cancer i.e., to pollution due to fossil fuels? We don't know what exactly caused the cancer, and we don't know how large the contribution of pollution was.
So a merely rational, sober calculation of risks and opportunities, pros and cons is impossible in practize. Especially in a liberal republic where actors equipped with strong resources (like energy companies and lobbies) are skewing public opinion.
Those scientists and administrators concerned with estimating risks rely on a formula: A risk is too high, when the product of event probability (likelihood of a meltdown, i.e.) and damage in case of the event, is too high. That means a particular procedure gets too risky, when either the event likelihood is low, but damage very high, or vice versa.
So, because a really encompassing, sober calculation is not possible, I say as a rule of thumb that nuclear power is clearly a case when event likelihood is low, but damage extremely high -- and the product too high to justify that technology, at least compared to regenerative energies. That's why I believe those should have priority. But until they can cover a sufficient share of the demand, I guess a certain share of nuclear energy is tolerable in contrast to mere fossil fuel use.