A previously developed car-to-car side impact model was further exercised in this study. This model consists of a vehicle with a deformable side structure representation, and an occupant simulating the SID dummy impact response. It has been demonstrated that the vehicle model correlates well with side impact test data. The occupant model has similar impact acceleration response as the SID dummy in the head, thorax and pelvis regions. In addition, good correlations were also found in the force-time histories of the thorax region when compared to cadaver drop tests. The model provided insights to the effects of various design parameters such as side structure stiffness and padding. Examining the side structure stiffness effect shows that there is a significant benefit for occupant protection by reducing the amount of intrusion until it is roughly equal to the initial distance between the occupant and the door inner panel. However, further stiffening the side structure would result in slight increase of occupant injury indicies. The model predicted that the use of padding on the door inner panel would reduce the occupant acceleration response but increase its deformation response. This prediction has been varified by two test series: (1). MVMA side impact test using Ford LTD and SID dummy, and (2). GM side impact test using a full size passenger vehicle and a Hybrid II dummy with the right arm removed. This result implies that, if the deformation response is also an important consideration in injury assessment, the use of additive padding needs to be accompanied by other design changes such as spacing or structure stiffness in order to compensate for this effect.