There is evidence to suggest that social isolation is associated with poor cognitive health, although findings are contradictory. One reason for inconsistency in reported findings may be a lack of consideration of underlying mechanisms that could influence this relationship. Cognitive reserve is a theoretical concept that may account for the role of social isolation and its association with cognitive outcomes in later life. Therefore, we aimed to examine the relationship between social isolation and cognition in later life, and to consider the role of cognitive reserve in this relationship. Baseline and two year follow-up data from the Cognitive Function and Ageing Study–Wales (CFAS-Wales) were analysed. Social isolation was assessed using the Lubben Social Network Scale-6 (LSNS-6), cognitive function was assessed using the Cambridge Cognitive Examination (CAMCOG), and cognitive reserve was assessed using a proxy measure of education, occupational complexity, and cognitive activity. Linear regression modelling was used to assess the relationship between social isolation and cognition. To assess the role of cognitive reserve in this relationship, moderation analysis was used to test for interaction effects. After controlling for age, gender, education, and physically limiting health conditions, social isolation was associated with cognitive function at baseline and two year follow-up. Cognitive reserve moderated this association longitudinally. Findings suggest that maintaining a socially active lifestyle in later life may enhance cognitive reserve and benefit cognitive function. This has important implications for interventions that may target social isolation to improve cognitive function.