Previous studies in humans interested in inhibitory synaptic activity at the level of the primary motor cortex (M1) have frequently used an electrophysiological technique call short intracortical inhibition (SICI). This technique consists of two subsequent pulses with transcranial magnetic stimulation (TMS), of which the first pulse (S1) has been argued to target local inhibitory connections, reducing corticospinal output generated by the second pulse (S2). However, the reduction of corticospinal output is not seen in every tested subject, sometimes S1 even increased corticospinal output. Thus there seems to be more occurring than just the targeting of local inhibitory connections by S1, indicating that the mechanisms of SICI are not fully understood. In the present study, in 18 healthy young subjects we applied a method allowing to segregate corticospinal volleys with different conduction times and investigated the effect of S1 on the excitation of these volleys. Our results revealed three major findings, which can be summarized as follows: i) S1 acted not only locally at the level of M1, but produced corticospinal activity which may influence corticospinal output by S2 at the spinal level; ii) there was not only reduced excitation of corticospinal volleys by S1, but also an increased excitation of corticospinal volleys with longer conduction times; iii) at the level of M1, S1 indeed targeted preferentially inhibitory synaptic connections, and these reduced corticospinal excitation of the fastest conduction corticospinal volleys. Our results indicate that underlying mechanisms in studies applying SICI should be interpreted with caution.