I second the microscope recommendation.
I'll also suggest some software that can help with reverse engineering, such as
https://server.ibfriedrich.com/wiki/ibfwikien/index.php/Reverse_Engineering which is a PCB Layout software. I own the software, but really haven't used it. it's personalized to you, so you can install it on as many machines as you would like, but you can only use it on one at a time. At the end of the year, they usually offer substantial upgrade discounts.
I tried to reverse engineer a small PCB manually and I had trouble. It was about 2x3cm and had mostly SMT components, but it had a few thru-hole components. A friend finally re-did it for me from photos and a BOM. His version made a lot more sense.
I have some other products that I want to reverse engineer:
a) A microwave control board
b) An antenna rotor control
Why the microwave. It's a large microwave/convection oven with probe. It does need to be fixed, but I want to add a new capability. I want to be able to set the probe temperature to lower than 110 deg. F. I'd like to be able to warm something cold to room temperature and 110F is too high for proofing yeast. So, I want to push a button and the temperature would be say 80 degrees less than the input temperature. When you pull the probe out, it would reset.
Why the antenna rotator. It's NLA and I'd like to add some web options. It looks like it's PIC based. Initially, I'd like to replicate the display on a web page and have a separate device do real-time position indicating. The actual position shown while turning is not correct.
It uses a form of Diseq/C over coax to control the rotor.
There would be decoding of the IR signals and the ability to send an IR signal.
Your going to need some image manipulation software because you will need to photograph both sides of the board at the same magnification and layer them with say 50% transparency. You can use that image in target.
So, you place the components, and draw the traces and it will spit out a schematic.
When laying out a PCB, footprints are difficult. Can't say what the best option is. PADS is a footprint generator.
X-rays are used by professional reverse engineering firms.
The panavise circuit board holder:
https://www.panavise.com/index.html?pageID=1&page=full&--eqskudatarq=26 Not sure how compatible it is with a microscope.
You might consider a color printer that can print at least 11" x 17" and probably large monitors.
Circuit specialists has an inexpensive line of soldering stations. I bought he Solderwerks BK5000 which is obsolete now, but similar to other products. I don;t like the de-soldering tool, but there is an upgraded version. I would reccomend getting a self-contained solder sucker. Cost could run <$300.00 USD. You can't use the iron and the desoldering tool at the same time. The desoldering tool would always hang-up in the holder. I had to buy a reamer to fix that. The tips keep slipping out of the soldering iron holder. Friction, with no lock. The soldering iron holder has an air extraction port for tip vapors.
If I had my druthers, METCAL probably has the best soldering irons. It's RF induction heating, so you really can't beat it/
Chip-quick has a removal alloy to help remove parts.
non-magnetic tweezers (titanium or Stainless Steel)
Your going to need specific tools to reverse engineer the CAN bus.