Synaptic transmission is transiently adjusted on a spike-by-spike basis, with the adjustments persisting from hundreds of milliseconds up to seconds. Such a short-term plasticity has been suggested to significantly augment the computational capabilities of neuronal networks by enhancing their dynamical repertoire. In this chapter, after reviewing the basic physiology of chemical synaptic transmission, we present a general framework-inspired by the quantal model-to build simple, yet quantitatively accurate models of repetitive synaptic transmission. We also discuss different methods to obtain estimates of the model's parameters from experimental recordings. Next, we show that, indeed, new dynamical regimes appear in the presence of short-term synaptic plasticity. In particular, model neuronal networks exhibit the co-existence of a stable fixed point and a stable limit cycle in the presence of short-term synaptic facilitation. It has been suggested that this dynamical regime is especially relevant in working memory processes. We provide, then, a short summary of the synaptic theory of working memory and discuss some of its specific predictions in the context of experiments. We conclude the chapter with a short outlook.