Each year influenza viruses are responsible for epidemic respiratory diseases with excess morbidity and mortality. The severity of influenza diseases ranges from mild upper respiratory tract infections to severe lower respiratory tract infections involving pneumonia, bronchiolitis and coincidental bacterial super-infections. The immune response to influenza viruses can be schematically divided into a cascade of non-specific and specific functions. These functions are involved at different well defined time points after infection. We describe in this manuscript three influenza models utilized in our laboratory: (i) a highly virulent influenza virus (influenza A/Hong Kong/8/68 (H3N2) virus) adapted to B6C3F1 mice, (ii) a mouse-adapted influenza A/Port Chalmers/1/73 (H3N2) virus, and (iii) a rat-adapted influenza virus (RAIV) model (influenza A/Port Chalmers/1/73 (H3N2)). This rat-adapted influenza model has been mainly utilized as a model to assess local immunotoxic effects of inhaled environmental pollutants such as phosgene. These host resistance models are also useful for assessing the effect of systemically-induced immunosuppression or immunomodulation by drugs or chemicals on the local pulmonary immune response to influenza virus. The comparison of these different models allowed two major conclusions: (a) viral replication and mortality are two different endpoints and are not necessarily linked (no mortality was observed with Port Chalmers virus in the mouse although the virus replicates to high titers in the lung with a kinetic pattern comparable to the one obtained with Hong Kong virus), (b) mortality, viral replication, and immune function assessment are different endpoints that can be used, depending on the question addressed.