Abstract : Sequential machines can be realized by series and parallel connections of components. These connections can be automatically learned by simulation, rating and selecting component models. Our goal here is threefold: (i) discuss implications for model reconstruction and reconfiguration in model engineering, (ii) present mathematical analyses of search simulation results, and (iii) conjecture that the whole search space of achievement components seems to be well ordered. This conjecture opens new perspectives for studying parallel and series connections of components upon a simulation-based framework. More generally, the automatic learning demonstrated here constitutes a formal basis for model reconstruction and reconfiguration in model engineering.