Models presented in the study suggest that the relationship between crash severity in urban highways and traffic variables same as traffic volume and flow speed, human factors, road, vehicle and weather conditions. Regarding that there was no model in previous works that could suggest the effects of all variables on crash severity of urban highways, models developed in current study, make it possible to investigate the effects of all the variables on crash severity, simultaneously. So, these models can be proper models in identifying the influential factors in crash severity. Thus, these models suggest that changes in crash severity doesn’t occur necessarily by any single dependent parameters, but occur as a simultaneous result of changes in these parameters.Results of the current study suggest that variables of driver age < 25, moving on with reverse gear, technical defect of the vehicle, vehicle collision with motorcycle and bicycle, bridge, frontal impact collisions, frontal-lateral collisions and multi-vehicle crashes in urban highways are among the main factors which usually increase the crash severity in urban highways.