Atmospheric modulation transfer function (MTF) is usually attributed to turbulence. The effect of aerosols are usually referred to as attenuation. However, recent experimental measurements of overall atmospheric MTF indicate significant difference between the turbulence and overall atmospheric MTFs, particularly when turbulence is weak such as in the early morning or late afternoon. We suggest here a physical explanation for those results which essentially relates to what we call a practical instrumentation-based atmospheric aerosol MTF which is a modification of the classical aerosol MTF theory. Based on radiative transfer theory, this practical approach takes into account the effect of finite field-of-view, finite dynamic sensitivity, and finite spatial bandwidth of every existing imaging system. These generally limit the scattering angles of received light to values far less than the diffraction limit for aerosols, thereby decreasing blur radius and increasing spatial frequency bandwidth. This can explain the broadening of the aerosol MTF from that theoretically expected. Also discussed is the asymptote value which the measured aerosol MTF approaches at high spatial frequencies, which is significantly higher than the theoretical prediction of turbid medium transmittance. Some important conclusions derived here are that the aerosol MTF cannot be referred to as constant attenuation, and in many cases it is the dominant part in the actual overall atmospheric MTF. In addition, there seems to be an inescapable trade-off between image resolution and image irradiance. This relationship can be determined by the system designer by considering field-of-view, and instrumentation dynamic range and spatial frequency bandwidth. He must choose between imaging of faint and bright objects at the expense of image quality, or imaging of either faint or bright objects with improved image quality. The concepts here are basic to all long-range imaging through the atmosphere.