To test whether long-term cortical adaptations occur bilaterally following chronic unilateral training with a simple motor task. Participants (n=34) were randomly allocated to a training or control groups. Only the former completed a 4-week maximal-intensity isometric training of the right first dorsal interosseus muscle through key pinching. Maximal strength was assessed bilaterally in four different movements progressively less similar to the training task: key, tip and tripod pinches, and handgrip. Transcranial magnetic stimulation was used to probe, in the left and right primary hand motor cortices, a number of standard tests of cortical excitability, including thresholds, intra-cortical inhibition and facilitation, transcallosal inhibition, and sensory-motor integration. Training increased strength in the trained hand, but only for the tasks specifically involving the trained muscle (key +8.5%; p<0.0005; tip +7.2%; p=0.02). However, the effect size was small and below the cutoff for meaningful change. Handgrip and tripod pinch were instead unaffected. There was a similar improvement in strength in the untrained hand, i.e., a cross-education effect (key +6.4%; p=0.02; tip +4.7%; p=0.007). Despite these changes in strength, no significant variation was observed in any of the neurophysiological parameters describing cortico-spinal and intra-cortical excitability, inter-hemispheric inhibition, and cortical sensory-motor integration. A 4-week maximal-intensity unilateral training induced bilaterally spatial- and task-specific strength gains, which were not associated to direct or crossed cortical adaptations. The observed long-term stability of neurophysiological parameters might result from homeostatic plasticity phenomena, aimed at restoring the physiological inter-hemispheric balance of neural activity levels perturbed by the exercise. ClinicalTrials.gov identifier NCT02010398.