The paper considers some physical aspects of the neutron space kinetics of critical and source-driven subcritical systems. The possibility of introducing some indicators to qualify the spatial nature of neutronic transients is investigated. It is shown theoretically and then proved by numerical examples that the separation of the eigenvalues of the mathematical operator defining the problem can be taken as a good indicator of the importance of space effects in time-dependent conditions. To obtain good physical insight into the phenomena, paradigmatically simple configurations are considered, and whenever possible, a fully analytical approach is used. The presented results evidence the limits of applicability of classic simplified models for transient analyses, such as point kinetics. In a second part, the paper considers the open problem of the choice of the weighting function to be used either for the generation of the kinetic parameters of pointlike models or for the exploitation of quasi-static procedures, analyzing comparatively the effect of different options on the results of transient calculations.