Basically, yes. While overall CPU load and likelihood of audio dropouts are significantly reduced, the total amount of data that needs to be processed is still the same. It’s just better distributed. So, while it’s possible to load more plug-ins and virtual instruments, the CPU and processing limits are still set by the available CPU performance, quality of drivers and audio I/O hardware used.