Therapeutic breakthroughs in neurological disorders have been hampered by the lack of accurate central nervous system (CNS) models. The development of these models allows the study of the disease onset/progression mechanisms and the preclinical evaluation of new therapeutics. This has traditionally relied on genetically engineered animal models that often diverge considerably from the human phenotype (developmental, anatomic, and physiological) and 2D in vitro cell models, which fail to recapitulate the characteristics of the target tissue (cell-cell and cell-matrix interactions, cell polarity, etc.). Recapitulation of CNS phenotypic and functional features in vitro requires the implementation of advanced culture strategies, such as 3D culture systems, which enable to mimic the in vivo structural and molecular complexity. Models based on differentiation of human neural stem cells (hNSC) in 3D cultures have great potential as complementary tools in preclinical research, bridging the gap between human clinical studies and animal models. The development of robust and scalable processes for the 3D differentiation of hNSC can improve the accuracy of early stage development in preclinical research. In this context, the use of software-controlled stirred-tank bioreactors (STB) provides an efficient technological platform for hNSC aggregation and differentiation. This system enables to monitor and control important physicochemical parameters for hNSC culture, such as dissolved oxygen. Importantly, the adoption of a perfusion operation mode allows a stable flow of nutrients and differentiation/neurotrophic factors, while clearing the toxic by-products. This contributes to a setting closer to the physiological, by mimicking the in vivo microenvironment. In this chapter, we address the technical requirements and procedures for the implementation of 3D differentiation strategies of hNSC, by operating STB under perfusion mode for long-term cultures. This strategy is suitable for the generation of human 3D neural in vitro models, which can be used to feed high-throughput screening platforms, contributing to expand the available in vitro tools for drug screening and toxicological studies.