The Gender Role Conflict Scale–Short Form (GRCS-SF) was derived from the widely used Gender Role Conflict Scale and is thought to measure men’s personal and relational distress from rigid adherence to restrictive masculinities. Extant research suggests the GRCS-SF may be best modeled as a second-order structure, consisting of 1 second-order gender role conflict (GRC) factor explaining the associations between four lower-order GRC domain factors. The present article revisited the factor structure of the GRCS-SF using confirmatory factor analysis in a large sample of college men (N = 1117). Contrary to previous research, a bifactor structure and a common factor structure evidenced better fit. Ancillary bifactor measures provided support for conceptualizing the theoretical construct of GRC and the dimensionality of the GRCS-SF instrument as being defined by four independent-yet-related first-order GRC factors. Furthermore, the restrictive emotionality and conflicts between work and family relations factors more consistently predicted unique variance in the 8 criterion variables (i.e., depression, anxiety, social anxiety, substance use, hostility, family distress, academic distress, and self-stigma of seeking psychological help) embedded in the nomological network of gender role conflict than did the success, power, and competition and restrictive affectionate behavior between men factors. Conclusions include: modeling the GRCS-SF using a correlated factors solution is recommended, use of the four raw GRC subscales scores may be permissible, and calculation of a raw GRC total score using all 16 items is contraindicated.