another 'very smort' question from fatty mc liver - could someone with experience in electronics give me some advice please?

most 8 bit cpus that can be had for money that makes sense (im saving that kidney for more complex designs that i maybe will take on later) have very little computational power and cannot do more than one thing at a time - is there anything else than fpga that could help me putting 2 8 bit cpus on a bus and, depending on software, help each other out with the task given? i understand that in most cases it will not yield any time savings and needs to be programmed to have positive effects;

what about the bus? it is my understanding that for each piece of information sent/received i loose a set amount of bus cycles due to architecture - can i save the time at least for the cpu? some actions might be just onesided data/signal transmissions - does it make sense to buffer some of that so that the cpu would sooner be free to do something else?

does it make sense to really have 2 busses and just pretend to the cpu that its 1 bus?

is memory mapping an option to improve performance?

i feel these machines can do a ton, but the architecture just sucked peanut butter slowing them down to the level we know - maybe im just not seeing straight?