Receptor tyrosine kinases couple a wide variety of extracellular cues to cellular responses. The class III subfamily comprises the platelet-derived growth factor receptor, c-Kit, Flt3 and c-Fms, all of which relay cell proliferation signals upon ligand binding. Accordingly, mutations in these proteins that confer ligand-independent activation are found in a subset of cancers. These mutations cluster in the juxtamembrane (JM) and catalytic tyrosine kinase domain (TKD) regions. In the case of acute myeloid leukemia (AML), the juxtamembrane (named ITD for internal tandem duplication) and TKD Flt3 mutants differ in their spectra of clinical outcomes. Although the mechanism of aberrant activation has been largely elucidated by biochemical and structural analyses of mutant kinases, the differences in disease presentation cannot be attributed to a change in substrate specificity or signaling strength of the catalytic domain. This review discusses the latest literature and presents a working model of differential Flt3 signaling based on mis-localized juxtamembrane autophosphorylation, to account for the disease variation. This will have bearing on therapeutic approaches in a complex disease such as AML, for which no efficacious drug yet exists.