135mm on a full-frame is actually a lot. If you shoot at f/8, your camera is looking out of an "entrance pupil" that has a diameter of 135mm/8 = 16.875mm. That means the far background will be blurred by a disk in the image that would cover 16.875mm in your focus plane. Which is a lot for an iPhone, less so for a car. So what about close background and foreground? Essentially you take that 16.875mm in the focus plane and scale it by the percentage of distance you are away from the focusing plane to figure out your blur disk diameter.
Let's try it on something of iPhone size (6cm×14cm). Settings (on a crop camera) at 71.5mm/4.8 which is only somewhat smaller than what you are working with. Turns out that the working distance for the following shot is about 40cm. So if we tilt the image so that going the length of the device of 14cm will change depth by 7cm (a 30° angle), we get 7cm/40cm=17.5% displacement, and 17.5% of 71.5mm/4.8 is about 2.6mm which we expect the blur disk to be at the bottom end of the device if the top of the device is in-focus.
The results appear to be in a similar ballpark as the handwaving applied here.
Now if you double the distance, the blur in relation to the subject becomes about half. The subject in the image also shrinks to about half which would additionally reduce blurriness, except that you'd crop until the subject again fills the frame. So that doesn't really count.
But in general, the recipe is sound: double the distance, half the blur in relation to the subject. However, if instead of cropping you adjust the focal length to account for the doubling of the distance while retaining your aperture number, the entrance pupil diameter also doubles, appearing at equal angular size from your subject as it did at the smaller distance. Which means that the blur in relation to your subject geometry remains the same.
So varying the distance while keeping the framing by also varying focal length does not really achieve a lot, while either increasing your aperture number or placing the subject in increasingly smaller portions of the image (and cropping afterwards, or by using a crop sensor in the first place) will help.
As a stupid rule of thumb: large subject wants large sensor area and aperture, small subject wants small sensor area and aperture.
Macro/closeup shooters almost invariably work with the smallest apertures their camera has available. Which may warrant making it artificially even smaller by stacking a teleextender between camera body and (macro) lens.
It depends on the size of your subjects on whether you already are in that "narrow aperture at all costs" part of the fight for depth of field or still outside that region where depth of field is the concern overriding every other image quality aspect.