Ageing is associated with a progressive decline in cognitive function, which occurs according to heterogeneous trajectories, dependent on multiple physiological and environmental components. To tackle this major challenge, we designed a project to test the effect of a tailored physical exercise intervention program in the cognitive function of a Portuguese elderly cohort, included in the AGA@4life project. The exercise program included aerobic and strength components, prescribed in a personalized approach according to the AGA@4life model, and implemented under direct control of two experienced professionals. The 33 included elderly participants were divided into two groups (intervention group –IG – and control group – CG) according to their willingness to participate in the physical training program. Cognitive function was evaluated with the Cambridge Neuropsychological Test Automated Battery (CANTAB) platform at baseline ant three-months after the intervention period in all the participants. The groups had similar clinical and demographic characteristics at baseline. After the intervention program, significant improvements in cognitive function were observed in the IG, but not in the CG. Particularly, a significant improvement in motor control, spatial working memory and visuospatial associate learning were depicted in the IG, which revealed an overall better cognitive performance as compared with the CG after the follow-up period. The results clearly identify physical exercise as an effective non-pharmacological tool to positively modulate age-related decline in cognitive function in older adults, particularly when prescribed in a personalized approach with a multicomponent structure as foreseen in the innovative AGA@4life model.