We have previously [Patuzzi and Rajan, Hear. Res. 60, 165-177, 1992] formulated a model to describe how the threshold elevations produced by a variety of independent, short-term cochlear manipulations add when the manipulations are combined. The manipulations were presumed to affect only the 'active process' in the cochlea. The present report applied this model to the effects observed after acute acoustic trauma in normal-hearing guinea pigs and in guinea pigs with idiopathic threshold losses. Successive loud pure-tone exposures were presented to the normal-hearing guinea pigs, while only a single exposure was presented to the guinea pigs with idiopathic hearing losses. Various parameters of exposure and inter-exposure delays were used to create a variety of threshold elevations, and the total hearing losses observed in the various groups were compared to the total hearing losses predicted by the model. In most cases a statistically-valid 1:1 relationship was obtained between the predicted values and the observed values. In cases where the model's predictions were found not to fit the data, this appeared to be due to inclusion of data previously defined to be outside the scope of the model. When such data were excluded, there was good agreement between the model's predictions and the observed data. The model was further tested by comparing its predictions with data obtained in studies of acute noise trauma in chinchillas and humans by other researchers. The model's predictions were found to agree with these data as well. Thus, across a number of different types and conditions of exposures, the model appears to provide a very good description of the additivity of threshold losses produced by acute acoustic trauma. The generality of and constraints on the model are discussed.