The problems with nuclear are mainly down to existing plant being an adaptation of Cold War weapon-making equipment. The PWR makes use of only a fraction of a percent of the energy in its fuel, and the fuel is then thrown away. That is why the waste is so long-lived. The fuel is largely unused. Also relevant to note that all of the major nuclear accidents except one were due to chemical rather than nuclear energy. Boiling water under pressure and zirconium fuel cladding are a bad combination, because once zirconium gets well alight it will burn underwater, or in steam. (The zirconium is necessary because it's one of the few structurally strong metals that doesn't capture neutrons.)
A radically different reactor design such as the molten salt LFTR would eliminate both the fire/explosion risk and pollution problems. We know this can be done, because it was tested in the 1950's and it worked. It could also be substantially cheaper to build once production is under way, because it would not need the massive safety precautions of the PWR. Since it would use nearly all of the energy in the fuel the amount of hazardous waste would be far smaller, and the world's nuclear fuel reserves would last many times longer.
As for fusion, it's criticized for taking so long to come to fruition, but when you consider that wind energy has been under development for nearly as long, and that global renewables expenditure is around half a trillion USD a year, whereas fusion research expenditure is a few billion.. I think you can see why that is.
The real unknown in this is LENR, which shows great promise although no-one has yet built a commercial product (Well, besides Rossi's alleged sale of a few units.) If LENR does turn out to be viable then it's basically game over for every other energy source. Again, research is woefully under-funded for what could be an invention as significant as the wheel. No, make that more significant. Part of the problem is that there is no textbook theory to explain it, and the academics have a credibility barrier when it comes to anything they can't first of all describe in maths, and only then in hardware.