Synthetic models, also called "surrogates", are commonly used in wound ballistics in order to simulate human tissues. Despite several surrogates are worldwide accepted and used; some of them have not been yet fully validated and their limits for forensic reconstructions have not been deeply investigated yet. In this work we present a homicide/suicide case involving three gunshots to the head with bullets retained in the skull or beneath the scalp. Reconstruction of these cases was performed preparing three individual synthetic head models based on post-mortem computed tomography (PMCT) measurements. Ballistic soap, polyurethane plates and 10% ballistic gelatine at 4°C were used as simulants in individually adapted thickness. Ballistic tests were performed using the questioned firearm and ammunition type. The damages on the synthetic models have been compared to the findings in PMCT and autopsy of the victims. Although the results highlighted general similarities in terms of injury characteristics, some of the experimental shots overpenetrated. Furthermore, the bullets recovered in the synthetic models did not show the same quality of deformations as the questioned bullets. This lack of bullet deformation in the synthetic models might be mainly attributed to the physical difference between real bones and polyurethane surrogate.