The use of general anaesthetics has facilitated great advantages in surgery within the last 150 years. General anaesthesia is composed of several components including analgesia, amnesia, hypnosis and immobility. To achieve these components, general anaesthetics have to act via multiple molecular targets at different anatomical sites in the central nervous system. Much of our current understanding of how anaesthetics work has been obtained within the last few years on the basis of genetic approaches, in particular knock-out or knock-in mice. Anaesthetic drugs can be grouped into volatile and intravenous anaesthetics according to their route of administration. Common volatile anaesthetics induce immobility via molecular targets in the spinal cord, including glycine receptors, GABAA receptors, glutamate receptors, and TREK-1 potassium channels. In contrast, intravenous anaesthetics cause immobility almost exclusively via GABAA receptors harbouring β3 subunits. Hypnosis is predominantly mediated by β3-subunit containing GABAA receptors in the brain, whereas β2 subunit containing receptors, which make up more than 50% of all GABAA receptors in the central nervous system, mediate sedation. At clinically relevant concentrations, ketamine and nitrous oxide block NMDA receptors. Unlike all other anaesthetics in clinical use they produce analgesia. Not only desired actions of anaesthetics, but also undesired side effects are linked to certain receptors. Respiratory depression involves β3 containing GABAA receptors whereas hypothermia is largely mediated by GABAA receptors containing β2 subunits. These recent insights into the clinically desired and undesired actions of anaesthetic agents provide new avenues for the design of drugs with an improved side-effect profile. Such agents would be especially beneficial for the treatment of newborn children, elderly patients and patients undergoing ambulatory surgery.