I know that many CPUs from that time use PLAs, in fact, my own CPUs use PLAs aswell! But there's something I don't understand: the arguments. For example, I have an ADD instruction with the following syntax (in bits):
FFFF AAA DDD BBB XXXX
(Here: F = function/opcode, the next are arguments: A = A source, B = B source, D = destination, X = not used)
But every time I look for the 6502 IS, I only find function matrix, no arguments. I want to know if the 6502 had arguments and if it did, how does it deal with them? I'm a newbie with the 6502 and I'm making a CPU inspired by the 4004, Z80 and 6502.

One or other of these links should help you:
http://www.masswerk.at/6502/6502_instruction_set.html
http://www.6502.org/tutorials/6502opcodes.html
http://www.obelisk.me.uk/6502/reference.html
https://www.dwheeler.com/6502/oneelkruns/asm1step.html

You will see that every instruction has a length: some are a single byte, some take one or two bytes of argument.

But because the 6502 is an accumulator machine, you won't find
ADD source1, source2, destination
because the destination for arithmetic is always the accumulator. And because it's an accumulator, it's also one of the sources. You will see
ADC operand
which will compute A <= A+operand, where operand may be interpreted as data, or an address, or the base for an indexed address, and so on.

Hope this helps.

(Oh, and welcome!)

Thanks, my dude