Yes it does. Imagine that instead of a train hitting a fly, a marble was hit by a baloon. The marble would push into the baloon as it decelerated, and eventually the marble and the part of the baloon with which it was in contact would stop moving relative to a stationary observer. During this process, of course, the marble and the relevant part of the baloon would be moving backwards relative to the motion of the baloon as a whole. Next, the baloon would return to its normal shape, the deformed part travelling in the same direction as the baloon as a whole, but with a slightly greater velocity. As the rigidity of the object involved increases, the magnitude and duration of the deformation decreases, but as no object can be totally rigid, the effect never disappears completely.