Simulation of vehicular communications is demonstrated to strongly depend on a plethora of settings which are often managed in a simplistic way. For instance heavy impact on bit-error-rate and packet-error-rate may come from physical layer phenomena which are either neglected - to make the overall simulation process faster - or managed by computationally heavy simulators - as in ray-tracing. As a result, current protocol simulations are not as much realistic as they could be. This paper proposes and evaluates two corrective models to the wireless simulation chain, and assesses their worthiness in terms of quantitative variations of the simulation outputs. An extensive comparative discussion on the results achieved with the corrective models provides a deeper insight into the validity and the limitations of simulations' outputs achieved. As a conclusion the paper shows that the proposed corrective models improve the significance of simulations and improve the significance of results.