Objectives: Existing research suggests there to be interrelated connections among physical activity, loneliness, and mental health. However, the possible mediating effects between them in middle-aged and older adults is unclear. This study aims to investigate the longitudinal relationship among physical activity, loneliness, and common mental disorders to clarify potential mediation effects. Methods: We analysed longitudinal data from 1973 participants aged 50 and over, drawn from the Longitudinal Ageing and Health Study Spain ("Edad con Salud", ageingandhealth.com). Loneliness was evaluated using the 3item UCLA loneliness scale, while physical activity was measured with the Global Physical Activity Questionnaire version 2 (GPAQ v2). Common mental disorders were assessed by considering lifetime episodes of major depressive and generalized anxiety disorders (MDD and GAD) following the criteria outlined by the Diagnostic and Statistical Manual of Mental Disorders (DSM-5). Employing a crosslagged panel model (CLPM), cross-lagged and autoregressive longitudinal associations among loneliness, common mental disorders, and physical activity were examined. Results: Physical inactivity predicted higher subsequent loneliness, which in turn predicted common mental disorders. The reverse pathways were statistically supported as well, suggesting bi-directional influences. All autoregressive associations were statistically significant. Discussion: The improvement of mental health through physical activity is facilitated by the mediation of loneliness, indicating the potential of the social aspect of physical activity as a preventive measure for MDD and GAD, as well as the need to address feelings of loneliness in individuals with common mental disorders in order to facilitate their participation in physical activities.