Microglia are the immune cells of the central nervous system (CNS). They patrol the brain environment with their ramifications and they respond quickly in the presence of pathogens and brain damages. Others and we have recently reported the existence of two different types of microglia, the resident and the newly differentiated microglia that are derived from the bone marrow stem cells. Of great interest is the fact that blood-derived microglial cells are associated with amyloid plaques and these cells are able to prevent the formation or eliminate the presence of amyloid deposits in mice that develop the major hallmark of Alzheimer's disease (AD). These cells are also recruited in the brain of other mouse models of brain diseases and acute injuries. They represent, therefore, a fantastic new vehicle for delivering key molecules to improve recovery, repair, and elimination of toxic proteins. However, recent studies have challenged this concept and raised concerns regarding the physiological relevance of bone-marrow-derived microglia. This review discusses both sides of the story and why the models used to follow the phenotypic fate of these cells are so crucial to reach the proper conclusion. Blood-derived progenitors have the ability to populate the CNS, especially during injuries and chronic diseases. However they do not do it in an efficient manner. Such a lack of proper recruitment may explain the delay in recovery and repair after acute damages and accumulation of toxic proteins in chronic brain diseases.