Objective Although there is emerging evidence for the factorial validity of the distinction between post‐traumatic stress disorder (PTSD) and complex PTSD (CPTSD) proposed in ICD‐11, such evidence has been predominantly based on using selected items from individual scales that describe these factors. We have attempted to address this gap in the literature by testing a range of alternative models of disorders of traumatic stress using a broader range of symptoms and standardized measures. Method Participants in this cross‐sectional study were a sample of individuals who were referred for psychological therapy to a National Health Service (NHS) trauma centre in Scotland (N = 195). Participants were recruited over a period of 18 months and completed measures of stressful life events, DSM‐5 PTSD, emotion dysregulation, self‐esteem and interpersonal difficulties. Results Overall, results indicate that a structural model incorporating six first‐order factors (re‐experiencing, avoidance of traumatic reminders, sense of threat, affective dysregulation, negative self‐concept and disturbances in relationships) and two second‐order factors (PTSD and disturbances in self‐organization [DSO]) was the best fitting. The model presented with good concurrent validity. Childhood trauma was found to be more strongly associated with DSO than with PTSD. Conclusion Our results are in support of the ICD‐11 proposals for PTSD and CPTSD.