Exercise has been found to influence molecular systems important for maintaining neural function and plasticity as well as treatment of neurologic disorders. The stimuli required to elicit plasticity are thought to be activity dependent. Several protocols of physical exercise have been used to explore its effects on brain function. However, it is becoming increasingly recognized that no single physical exercise model is likely to fulfill all therapeutic needs. Varied interpretations of data derived from animal models have given rise to the lack of uniformity in the description and control of various features of the physical exercise stimulus, ranging from low to high intensity, intermittent to sustained, short to long durations, and different modes of activity. This article first describes the characteristics of the most frequently used animal models and goes on to review brain plasticity in intact animals and the usefulness of these models for the study of brain disorders. In this regard, animal models that investigate the beneficial effects of exercise on the brain before and after brain injury are discussed. A challenge for future studies is to better evaluate the usefulness of physical exercise protocols for preventing or treating brain disorders.