There traits are not all essential, but tend to be present in most doers of great things in science. First, successful people exhibit more activity, more energy, than most people do. They look more places, they work harder, they think longer than less successful people. Knowledge and ability are much like compound interest -- the more you do the more you can do, and the more the opportunities are open for you. Thus, among other things, it was Feynmann's energy and his constantly trying new things that made one think he would succeed.
This trait must be coupled with emotional commitment. Perhaps the ablest mathematician I have watched up close seldom, if ever, seemed to care deeply about the problem he was working on. He has done great deal of first class work, but not of the highest quality. Deep emotional commitment seems to be necessary for success. The reason is obvious. The emotional commitment keeps you thinking about the problem morning, noon and night, and that tends to beat out mere ability.
You need a vision of who you are and where your field is going. A suitable parable is that of the drunken sailor. He staggers one way and then the other with independent, random steps. In $n$ steps he will be, on the average, about $\sqrt{n}$ steps away from where he started. but if there is a pretty girl in one direction he will get a distance proportional to $n$. The difference, over a life time of choices, between $\sqrt{n}$ and $n$ is very large and represents the difference between having no vision and having a vision. The particular vision you have is less important than just having one - there are many paths to success. Therefore, it is wise to have a vision of what you may become, of where you want to go, as well as how to get there. No vision, not much chance of doing great work; with a vision you have a good chance.
I must now take up the unpleasant topic of selling your ideas. Too many scientists think that this is beneath them, that the world is waiting for their great results. In truth, the other researchers are busy with their own work. You must present your results so that they will stop their own work and listen to you. Presentation comes in three forms: published papers, prepared talks, and impromptu situations. You must master all three forms.