Vaccination is perhaps the most beneficial public health tool in history. The battle has been won for smallpox, diphtheria, polio, measles, yellow fever and several other diseases (1), saving millions of lives and changing the demographics of the world. However, the war is not over and many challenges such as HIV, influenza, hepatitis C virus (HCV), Mycobacterium tuberculosis, Malaria and others remain. Most vaccines have been developed empirically and despite their success we understand little about the ultimate mechanism of their protective immunity. Although the majority of concepts of the new vaccine approaches stem from in vitro data, major insights still have to come from in vivo analysis. The past decade has witnessed a significant progress in elucidating mechanisms of vaccine-induced immune response (2, 3), providing data necessary for rational development and design of 'smart' vaccines. The majority of current vaccines rely on protective immunity induced by infection with live, but attenuated pathogens. However, in most of these approaches there is little or no external control over the process of mounting the immune response. Depending on the protective principle needed for controlling certain pathogens, different types of immune response may be desirable, including the production of protective antibodies and various components of specific cellular immune response. In addition, for successful protection against many pathogens, it is essential that immune response mechanisms are operative at the site of infection, e.g. mucosal tissue as the most frequent site of pathogen entry. Novel approaches of vaccine development include delivery of the gene of interest and expression of antigens within a host. These approaches provide much better possibilities to control the process of inducing and maintaining a specific immune response, appropriate to best counteract particular pathogen. The genes encoding the antigens may be delivered to the host by introducing DNA, RNA, modified bacteria or virus vectors. Also, synthetic peptides based vaccines and peptide-loaded dendritic cells (DCs) show promising results as another modality of vaccine approaches. In this review we briefly discuss why traditional vaccine approaches were not so effective against several present-day epidemic diseases. Thereafter, we highlight some immunological principles that give foundation to viral vector-driven vaccines. We put a special emphasis on how the knowledge gained from cytomegalovirus can be used to design more efficient vaccines and vaccine vectors against multiple human pathogens.