@layman,
Whatever. SR was developed to explain one particular experiment, and it does that well. However, it's only applicable in limited circumstances, e.g. when gravity can be safely neglected, and on the 'local' level. When one cannot neglect gravity and/or when a diachronic explanation of a phenomenon is required -- that is to say, to explain any real-life astronomical phenomenon extending over some non-negligible length of time -- then GR is required. The problem is that we don't know how to solve GR equations (Einstein field equation) under most circumstances.... That's why SR is still around: its equations are simple enough, and many GR situations can be approximated through SR equations, by neglecting this and that variable, so scientists still use SR when they think they can afford to.
Same issue applies to quantum mechanics: the Schrödinger equation is supposed to describe any quantum system, including therefore the entire universe. It's the equation that rules the world... The problem is we don't know how to solve it other than for the simplest of systems (one atom of hydrogen or helium), and even then, only through some simplifications and approximations.
In fact, the same issue applied to Newtonian physics, that are far from obvious to compute once you go beyond three celestial objects. We would be hard-pressed to predict the trajectories of, say, 5 or 6 objects of similar size interacting with one another through Newtonian gravity, even if we knew everything we needed to know about their initial position, speed, spin, etc. The calculations are just beyond our present grasp of the mathematics involved. We can calculate the trajectories of planets in the solar system because the size of the sun respective to the planets' is so large that one can safely neglect the gravitational pulls exerted by the planets on the sun, thus simplifying the equations quite a lot.