Neurodevelopmental disorders are challenging to study in the laboratory, and despite a large investment, few novel treatments have been developed in the last decade. While animal models have been valuable in elucidating disease mechanisms and in providing insights into the function of specific genes, the predictive validity of preclinical models to test potential therapies has been questioned. In the last two decades, diverse new murine models of Down syndrome (DS) have been developed and numerous studies have demonstrated neurobiological alterations that could be responsible for the cognitive and behavioral phenotypes found in this syndrome. In many cases, similar alterations were found in murine models and in individuals with DS, although several phenotypes shown in animals have yet not been confirmed in the human condition. Some of the neurobiological alterations observed in mice have been proposed to account for their changes in cognition and behavior, and have received special attention because of being putative therapeutic targets. Those include increased oxidative stress, altered neurogenesis, overexpression of the Dyrk1A gene, GABA-mediated overinhibition and Alzheimer's disease-related neurodegeneration. Subsequently, different laboratories have tested the efficacy of pharmacotherapies targeting these alterations. Unfortunately, animal models are limited in their ability to mimic the extremely complex process of human neurodevelopment and neuropathology. Therefore, the safety and efficacy identified in animal studies are not always translated to humans, and most of the drugs tested have not demonstrated any positive effect or very limited efficacy in clinical trials. Despite their limitations, though, animal trials give us extremely valuable information for developing and testing drugs for human use that cannot be obtained from molecular or cellular experiments alone. This chapter reviews some of these therapeutic approaches and discusses some reasons that could account for the discrepancy between the findings in mouse models of DS and in humans, including: (i) the incomplete resemble of the genetic alterations of available mouse models of DS and human trisomy 21, (ii) the lack of evidence that some of the phenotypic alterations found in mice (e.g., GABA-mediated overinhibition, and alterations in adult neurogenesis) are also present in DS individuals, and (iii) the inaccuracy and/or inadequacy of the methods used in clinical trials to detect changes in the cognitive and behavioral functions of people with DS. Despite the shortcomings of animal models, animal experimentation remains an invaluable tool in developing drugs. Thus, we will also discuss how to increase predictive validity of mouse models.