Objectives: To assess whether social capital benefits older adults’ self-rated health and well-being and whether physical activity mediates this relation. Methods: A survey study was conducted among members of a sociocultural organization (age ≥55 years), both cross-sectionally (baseline Time 1; N = 959) and longitudinally (3-year follow-up Time 2; N = 409). Results: Specific indicators of social capital were positively, though modestly, related to health and well-being at Time 1 and Time 2. Experienced connectedness with age peers emerged as the strongest predictor. Physical activity only mediated the relation with experienced safety in society. Discussion: The relative importance of older adults’ experienced connectedness with their age peers underlines the importance of internalized group membership as a determinant of their health and well-being. Physical activity seems to play only a minor mediating role.