To study Alzheimer's disease (AD), a variety of mouse models has been generated through the overexpression of the amyloid precursor protein and/or the presenilins harboring one or several mutations found in familial AD. With aging, these mice develop several lesions similar to those of AD, including diffuse and neuritic amyloid deposits, cerebral amyloid angiopathy, dystrophic neurites and synapses, and amyloid-associated neuroinflammation. Other characteristics of AD, such as neurofibrillary tangles and nerve cell loss, are not satisfactorily reproduced in these models. Mouse models that recapitulate only specific aspects of AD pathogenesis are of great advantage when deciphering the complexity of the disease and can contribute substantially to diagnostic and therapeutic innovations. Incomplete mouse models have been key to the development of Abeta42-targeted therapies, as well as to the current understanding of the interrelationship between cerebral beta-amyloidosis and tau neurofibrillary lesions, and are currently being used to develop novel diagnostic agents for in vivo imaging.