Current search for a new effective vaccine against tuberculosis involves selected antigens, vectors and adjuvants. These are being evaluated usually by their booster inoculation following priming with Bacillus Calmette-Guerin. The purpose of this article is to point out, that despite being attenuated of virulence, priming with BCG may still involve immune mechanisms, which are not favourable for protection against active disease. It is postulated, that the responsible ‘decoy’ constituents selected during the evolution of pathogenic tubercle bacilli may be involved in the evasion from bactericidal host resistance and stimulate immune responses of a cytokine phenotype, which lead to the transition from latent closed granulomas to reactivation with infectious lung cavities. The decoy mechanisms appear as favourable for most infected subjects but leading in a minority of cases to pathology which can effectively transmit the infection. It is proposed that construction and development of new vaccine candidates could benefit from avoiding decoy-type immune mechanisms.