@Blagtastic88:
Personally, I've found that molten salt reactors solve many of the problems of traditional light water reactors and most other reactors that operate under pressurized conditions. They have a proof of concept in Oak Ridge experiment. Materials have improved since then to further solve any corrosion problems and with further advances in turbine technologies, closed cycle turbines could very efficiently produce energy from these reactors. Also, fuel and waste can be added or removed on-line with molten salt reactors. This is especially true in the case of Thorium breeder reactors in which you can bubble out your fissile Uranium from your fertile Thorium using Florine gas. Also, by exploiting the Thorium fuel cycle they have lots of fertile material to turn into fissile material as well as many useful decay products that could be extracted other industrial uses. Thorium also produces fewer transuranics. Just my opinion, but I think that molten salt reactors could be a viable alternative to current coal plants and help not only cut back on carbon emissions, but create a stable domestic fuel economy for electricity production.
Also, fusion is really just a too far off to practical in solving current energy problem. Solar is too inefficient even as more multi-junction cells are produced. The Artificial Leaf was a cool idea, but Hydrogen is a pain to store and has terrible energy density. Though with the materials involved, it could be a novel concept for water purification and individual scale power supplementation.
I don't know if any nuclear energy source involving fission is really a good idea for use on or around planet Earth. I understand that a lot of natural fission reactions are occurring around and under us all the time, but I think it's best that we keep their rates about where they currently are.
As you've pointed out, fission reactors create lots of random radioactive byproducts. Sure, some can be useful, but most aren't, and you're never going to find a fission reaction that's perfectly clean/predictable. With our current level of technology, there really is no way to guarantee prevention of "bad" byproducts. And I mean really bad, as in having radioactive half lives on the order of tens of thousands of years or more. Is there something about the molten salt reactor that I don't know that eliminates these things? If not, then I think that it's probably not the best idea for the environment, even if it significantly reduces carbon emissions.
The huge advantage of the fusion reactor is that you always know exactly what material it is creating, how radioactive it is, how harmful it is for the environment, etc. Fusion's awesome because of its predictability. While most modern fusion schemes make use of copious neutron production, it's also quite easy to create neutron shields out of materials that will also be relatively safe when irradiated. Designed properly, a fusion reactor would at worst simply irradiate itself (the containment vessel), which you could easily build out of radiation hardened materials that would be reusable after allowing to sit for a couple 20 year half lives inside a concrete box.
And let's not forget that a typical fission reactor needs like 100 years worth of fuel in its belly just to get to critical mass, which is what makes meltdowns such a problem in the first place. On the other hand, a fusion reactor works best when there is very, very little fuel in it at any given time. Of course, we still haven't made a working fusion power plant, so it's kind of silly for me to be making fusion sound so great in comparison to existing nuclear energy schemes.
Ultimately, I think we're much better off pursuing things like solar, wind, wave, and geothermal power in the future. Those, and energy storage devices (like better batteries and fuel cells). On this front, carbon nanotubes are really making me optimistic about future technologies. It's funny that you mention the horrible energy density of hydrogen: When we run out of hydrocarbon fuels, people are going to be really sad to find that we simply cannot fly planes anymore. The whole reason it's possible to design a plane/helicopter is because of how compact the carbon bonds make the main hydrogen fuel source. But what's really cool is that some research has shown that carbon nanotubes may bond so well to molecular hydrogen when heated that they can be used as fuel cells with energy densities possibly exceeding that of modern jet fuel. They can also be used to make batteries, quantum computers, and even room temperature light bulbs. I'll say that again: Room. Temperature. Lightbulbs. Think of the energy savings!
Someone stop me now before I start typing about space elevators and shit. Invest in carbon, people.
edit: Oh yeah, you can make flexible solar cells out of them, too.