How do physicists develop theories? I'm wondering about the way physicists develop "unified theories" and models about the universe, like string theory, or quantum mechanics. People (on this forum for example) often say things like "at the current state of the development of string theory...", as if predicting further development of the theory, or rather as if understating that such theories are continuously developed.
My questions are : how can "unified theories" be incomplete and need to be developed further if they are trusted by scientists to be true? How does development of such theories occur, through continuous small progress or rather through rare epiphanies (eurêka)?
Sorry for somewhat vague title question, but the subject of the question is to me quite vague as well --- my studies are rather far from hardcore theoretical physics. I'd just like to understand better how research is done in physics.
 A: (I presume that you meant "Physicist" not "Physician" as in one who practices medicine)
Scientific theories are usually a group of ideas that are linked either through a common phenomenon or a common method.  Usually, for more developed theories, these are in the form of postulates(think of quantum mechanics or statistical mechanics), with other ideas developed from them.  For newer theories (such as string theory) these postulates are still being figured out (and continually tested against experiments).  As far as regarding them as "true", in some cases it means that it fits mathematically with the rest of known physics(if there are no tests able to yet be done in the field), or fits experimental evidence better than other theories do.
And all fields of science have continous improvement.  Even the theory of classical mechanics is still being developed mathematically.  
The development of these theories is small incremental progress in the right direction.  It usually is that someone notices that the current theories have some problem, and multiple theories (most of them wrong) are proposed to correct it, then someone (or a group of people) notice a way to unify the evidence in a new way that is better than all of the others.  A great example is the development of Relativity and Quantum Mechanics in the early 20th century.       
