Cooperative adaptive cruise control (CACC) has drawn wide attention in recent years for its potential throughput benefit, as it is a promising intermediate technology to the highly connected and automated vehicles. The impact of CACC on multilane highways has been the subject of several studies, but they assumed traffic under a uniform speed limit. Recent research has revealed that traffic performs differently under a differentiated per-lane speed limit (DPLSL) policy with heavy vehicle (HV) restricted lanes. Whether the benefits of CACC still remain under a DPLSL policy has not been explored. This study developed cellular automaton models to incorporate CACC-equipped and non-equipped vehicles (i.e., passenger cars, HVs) on a two-way eight-lane highway with a DPLSL. Results shown throughputs by lane increase up to 78.5% as the CACC car market penetration rate (MPR) rises. Such increases became sharper (i.e., ≥10%) for inner lanes (i.e., HV restricted lanes) and outer lanes after reaching a 40% and a 60% CACC car MPR, separately. Moreover, HVs induced a 1.5% to 15.7% throughput reduction across lanes even under higher CACC car MPRs (i.e., 60%, 80%). This DPLSL policy may cause the lanes to experience a throughput penalty when they are adjacent to lanes with a different speed limit as the MPR of CACC cars rises. Lastly, in traffic with a 60% CACC car MPR, increases are brought further by considering 10% of HV with CACC, especially on those HV non-restricted lanes. The study is helpful for policy makers to further prepare for the prevalence of CACC in the forthcoming years.