A point-neuron model for the activity of individual cells in the medial superior olive (MSO) is described and shown to generate discharge patterns consistent with the activity of real neurons as reported in response to low-frequency sinusoidal stimulation. Inputs to the model cell are specified as primarylike firing patterns, and the cell membrane characteristics are specified in terms of constant-potential sources and time-varying conductances. Some conductances are determined in response to the input firings and some in response to output firing times, which are generated when the membrane potential of the model cell crosses threshold. Output firing patterns generated by the model cells are consistent with those reported from neurons in dog and cat MSO. These patterns are also compatible with those of the functionally specified coincidence model described in Colburn et al. (1990). Given these observations, the following questions are addressed: What parameter values in the point-neuron model are required to generate output patterns like those observed? How do these values compare to those expected or estimated from intracellular measurements in brainstem neurons? How might one reconcile the fact that inhibitory inputs are not necessary in the model for the generation of the observed firing patterns with the fact that MSO cells receive numerous inhibitory inputs?