Air filters are a bit of a con. A guy I know who manufactures performance parts for Subaru's did some testing on his own drag racing car which at that time was well over 500bhp, and ended up over 600. The tests were on a variety of panels and cones from several manufacturers, and conducted on the same car, same rolling road, and over two consecutive days, one quite cool and damp, the other slightly warmer and drier. Types tested included the OE paper style panel, wire gauze and cloth such as K&N and Green, and foam from Jetex and PiperX. The results showed a spread of just ±2bhp. Intake pressure and temperatures confirmed these findings with very little benefit being found for using any of the expensive aftermarket replacements. The only plausible justification drawn was a financial gain if you intend to keep the car a while, and by using a 'lifetime' item you may eventually save a few quid over the serviceable OE item.
A little less scientific was his method of checking filtration, whereby he placed each filter in a clean tin and deposited the same handful of dirt from his vacuum cleaner on them, then placed it on top of an engine on tickover. Visually he recorded that the OE item appeared to filter the best, letting no large particles through, and negligible amounts of very fine dust. The gauze filters were almost as good, but the foam filters were awful letting lots of larger particles through.
Exhausts can indeed make more of an advantage, especially to a turbo-charged car. The faster and free-er the gas flows, the faster the turbo will spin, increasing boost pressure, and speed of spool-up. However, it's not always as simple as that as modern exhausts have a lot of obstacles. Catalytic converters can spoil the party greatly, so getting rid of them is always a good idea, apart from when you need to MOT the car or are stopped for a VOSA check which could cause legal issues for road use. Sports cats which flow free-er are good here, and often cars have several cats when one is enough. My own Impreza had 3, but one decent sports cat would still have seen it pass an MOT. I had a package on mine that removed the mid section cat, and if I had kept it I planned to remove the up-pipe pre-cat, and replace the down-pipe cat with a sports variant. The ECU can adjust a little to cope with this, but a remap would be required to maximise the potential of such a mod. But it goes even further than that, as the amount of boost the car would be able to achieve from that would see me overwhelm the fuel pump and injectors, forcing more mods and expense to make sure I both got the maximum, AND that the engine ran properly. Running lean on a Subaru can cause fatal engine damage. It's also highly likely that after all that that you will be nearing the boost limits of the turbo, so that starts the whole mod/paying process all over again.
On non-turbo cars the benefits are much lower, and in truth, most people buy them to get a sexy chrome tailpipe and some extra farty noises. Reducing back pressure on ordinary, low powered engines will liberate very little performance, and may cause issues with valve movement too. Car companies plough millions into setting their cars up for a mix of reliability, economy and performance; aftermarket exhaust makers do not, and unless you know a little about cars then again, you could be wasting money.
Wheels can indeed make a difference, but again, you must do your homework. Generally, people buy alloys for cosmetic reasons, no performance. There can be some benefit to the usually wider tyre providing more grip, and lower profile reducing sidewall flex to give a shaper feel. People often go too far though, with stupidly wide tyres which become more prone to aquaplaning in the wet, far outstrip the grip requirements of the car, and can ruin the driving feel. Skinny profiles can remove the feel of grip where a driver senses nearing the limits, so if/when they do run out of grip/talent, it is sudden and unrecoverable.
Alloys are also often no lighter than their steel counterparts. As steel is an extremely strong metal, it doesn't take much to give the structural integrity required for the job in hand, whereas alloy is lighter than steel, but also not as strong, so more has to be used to achieve the structural integrity of the wheel. Different alloys are lighter (magnesium especially), but with that comes even more expense. Flow-forming reduces oxygen content of the alloy making it stronger, so less can be used and weight reduced, but again, cost is a factor. OZ's Ultra/Super Leggera range were a decent flow formed wheel at a reasonable price. I fitted some to one car and they did indeed improve both handling and performance. Performance comes from the engine needing less of its power to rotate the wheel, leaving more to actually accelerate the car along. Think of a bicycle wheel which weighs almost nothing and is easy to spin, compared to a heavy car wheel which takes a lot of force to get moving. Similarly, the handling benefit comes from the springs and dampers having to work less to recover from movement. Imagine holding a bucket full of water at arms length and moving it up and down. Your arms are damping the movement and recovering the position, but with all the weight of the water it is hard and slow to do. Now pour the water away and raise/lower the bucket again. With almost no weight it is easy.