This study explores the dual modulation of monoamine oxidase B (MAO-B) and the Sigma-1 receptor (Sig-1R) as a promising multitarget strategy for central nervous system drug discovery. Robust QSAR-based machine learning models were developed using curated ChEMBL datasets, achieving high predictive performance and revealing key structural features associated with dual activity. Virtual screening of natural compounds identified candidates with multitarget potential, which were further validated by molecular docking and molecular dynamics simulations, establishing general SAR principles to support rational molecular design and lead prioritization in multitarget medicinal chemistry.