In order to validate the model of hearing hazard from intense sounds [G. R. Price and J. T. Kalb, J. Acoust. Soc. Am. 90, 219--227 (1991)], hearing loss data have been sought from impulse noise exposures which extend the range of exposures already examined with the model. One such impulse (reported in Price et al. at this meeting) is that produced by air bag deployment, which contains a mixture of high-frequency energy often superimposed on a low-frequency pedestal, reaching peak pressures near 170 dB and with B durations of 150 ms. The model's predictions for firearms impulses in the same intensity region correlated well with actual threshold shifts [Price et al., J. Acoust. Soc. Am. 97, 3343 (1995)]; however, the hazards calculated with the air bag impulses were generally too large. Examination of the data suggested that middle-ear muscle activity needs to be accounted for, that corrections for the angle of incidence of the wavefront are needed, and that the existing model included too much damping for large displacements of the middle ear.