Over the years we have gathered expertise on A- and B-site substituted perovskite-type oxides used for energy production processes and environment protection. We are interested to exploit the possibility to segregate/exsolve transition (e.g. Cu, Ni) and noble metals (e.g. Pd, Rh, Pt) out of the lattice of perovskite-type mixed oxides of various composition to obtain active, stable and selective catalysts. Using state-of-the-art advanced characterization methods we look for the boundary experimental conditions that enable to maximize the fraction of segregated metal and the fraction that can be regenerated within the lattice upon oxidation. This approach is required to define the reactive conditions under which the activity and stability of the catalyst can be optimized. While it is an extremely intriguing material property, likely not limited to this class of materials, the definition of the so-called self-regeneration needs to be revisited and verified for every perovskite-type composition and active metal.