A widely accepted 'Kiruna conjecture' (ICS-1,1992) confirmed the importance of the Magnetic Reconnection (MR) in substorms but expressed doubts that the MR plays any role at substorm onset initiated in the near tail (7-12 Re). Inspired by recent modelling results and observations of singular Thin Current Sheets (TCS) in the near tail, we return again to the MR as possible general mechanism responsible for substorm onset, for the following activations as well as for the pseudobreakups proceeding this onset. We summarize recent observational evidence of the MR occurring in the near tail and discuss some predictions of reconnection theory concerning the reconnection rate amplitude, initiation of substorm onset in the near tail and low probability to observe fast tailward flows. Importance to modify the view of reconnection to include the multiple impulsive MR events confined in the very small (and thin) portions of the near-Earth plasma sheet is emphasized, and possible controlling factors are discussed. The impulsive magnetic reconnection may influence the plasma sheet dynamics by producing some specific phenomena, like (1) thin current sheets, (2) small-scale plasmoids (or flux ropes) containing fast bulk flous, (3) bifurcation of the tail current etc. We suggest a few more tests to distinguish the reconnection signatures at substorm onset and discuss the importance of small cross-tail size of the dissipative process. We conclude that magnetic reconnection is a very probable mechanism to initiate the substorm onset, however the reconnection-based scenario should also include the coupling of this meso-scale process with other large-scale and micro-scale processes to provide a realistic description of substorm dynamics.