A 2-mm-thick plastic shell with 5-10-mm-thick tapered plastic spacer in the shape of a "pinna"-specific anthropomorphic mannequin (SAM) head model is being used for determination of the specific absorption rate (SAR) of cellular telephones for compliance testing against IEEE and the International Commission on Non-Ionizing Radiation Protection (ICNIRP) Safety Guidelines used in the U.S. and Europe, respectively. We have used three-dimensional computer-aided design files of the SAM Model with 1-mm resolution to calculate peak 1- and 10-g SAR for "cheek" and "15°-tilted" positions of some typical telephones for comparison with those for three anatomic models of the head to show that the SAR obtained for SAM is up to two or more times smaller than for anatomic models. This is due to the shift of the high SAR locations to a low radiated fields region away from the antenna, particularly at 835 MHz, and a substantial physical separation from the absorptive phantom at 1900 MHz. Due to the use of lossless plastic for the "pinna," another handicap of the SAM model is the total lack of knowledge of 1- or 10-g SAR in the pinna tissues required by all safety guidelines (current or proposed). To remedy this situation, we propose a modified SAM with a lossy "pinna," for which 1- and 10-g SARs are relatively close to those for anatomic models, provided we use a fluid of higher conductivity than that currently used for compliance testing at 835 MHz. Lastly, we compare the implications of the current IEEE and ICNIRP guidelines and the newly proposed IEEE guidelines with a relaxed limit of 4.0 W/kg for any 10-g of tissue of the pinna for maximum allowable powers for cellular telephones at 835 and 1900 MHz to show that the newly proposed relaxed IEEE limits will allow radiated powers that may be 8-16 times those permitted by the current IEEE Standard and up to two times higher than those permitted under ICNIRP guidelines used in over 30 countries.