With the significant increase of life expectancy of populations in societies today, the importance of the discovery of drugs associated with neurodegenerative diseases has emerged. Therefore, neurodegenerative diseases are an important topic in Medicinal Chemistry. Although drug discovery is considered a complex and slow process, new approaches and methods have been developed with the intention of finding new chemical entities in more efficient ways. This work provides a review of virtual methodologies applied in drug discovery and especially a new model for the prediction of MAO-A inhibitors using a multi-target QSAR methodology. This model involves a mixed approach containing simple descriptors based on atom-centered fragments and functional groups (DRAGON) and topological substructural molecular design descriptors (MODESLAB). This unified multi-species QSAR model was validated through a virtual screening of a new series of oxoisoaporphine derivatives, taking into account the information in the calculated fragmental contributions. Therefore, this method represents a useful tool for the in silico screening of MAO-A inhibitors.