Carbon nanotube (CNT) fiber electrodes have demonstrated exceptional spatial selectivity and sustained reliability in the context of intrafascicular electrical stimulation, as evidenced through rigorous animal experimentation. A significant presence of unmyelinated C fibers, known to induce uncomfortable somatosensory experiences, exists within peripheral nerves. This presence poses a considerable challenge to the excitation of myelinated Aβ fibers, which are crucial for tactile sensation. To achieve nuanced tactile sensory feedback utilizing CNT fiber electrodes, the selective stimulation of Aβ sensory afferents emerges as a critical factor. In confronting this challenge, the present investigation sought to refine and apply a rat sciatic-nerve model leveraging the capabilities of the COMSOL-NEURON framework. This approach enables a systematic evaluation of the influence exerted by stimulation parameters and electrode geometry on the activation dynamics of both myelinated Aβ and unmyelinated C fibers. The findings advocate for the utilization of current pulses featuring a pulse width of 600 μs, alongside the deployment of CNT fibers characterized by a diminutive diameter of 10 μm, with an exclusively exposed cross-sectional area, to facilitate reduced activation current thresholds. Comparative analysis under monopolar and bipolar electrical stimulation conditions revealed proximate activation thresholds, albeit with bipolar stimulation exhibiting superior fiber selectivity relative to its monopolar counterpart. Concerning pulse waveform characteristics, the adoption of an anodic-first biphasic stimulation modality is favored, taking into account the dual criteria of activation threshold and fiber selectivity optimization. Consequently, this investigation furnishes an efficacious stimulation paradigm for the selective activation of touch-related nerve fibers, alongside provisioning a comprehensive theoretical foundation for the realization of natural tactile feedback within the domain of prosthetic hand applications.