The moon is said to have significant amounts of Helium 3 because it's not protected from cosmic rays. He-3 plus Deuterium (H-2) needs little energy to start a fusion into He-4 plus a neutron (which decays into H-1 plus an electron with a half time of about 10 minutes). The other "easy" fusion is Tritium (H-3) plus Deuterium, but Tritium is radioactive with a half-time of 12 years (rather weakly radioactive per nucleus... but still enough to kill you if unprotected) and naturally available in small traces only (through cosmic rays only, IIRC). You would have to produce it artificially. 2x H-2 -> He-4 might be within our grasp as well.
The holy grail is fusing 4x H-1 into He-4 which would produce by far the most energy (that's what happens in the core of the sun), but that's currently out of what we can achieve.
Here's a graph of the average binding energy per nucleon (i.e. proton or neutron) for nuclei (i.e how much energy per nucleon would be needed to rip the nucleus apart into neutrons and protons):
To figure out how much energy is reased on a fusion, choose two or more source elements, add up their binding energies for all of its nucleons and compare it with the sum of binding energies of all the nucleons of one or more target element into which they can fuse. If the source has less, then energy would be released on a fusion. It's pretty much the same for fission, but our fission reactors use elements from the far right of the curve, with comparingly pathetic binding energy differences.
That's not all of it, however. What's not included in the graph is that you also have to produce enough energy to make it happen (i.e. to cause conditions at which the strong nuclear force overcomes the repulsion of the electric charges) and that's the challenge of it all.