I've read DI will cover up to 2,5 times the underboat's depth (in wide mode), so seeing 15m on the side when there is 5m underboat doesn't seem too disturbing.
On your screenshot the object displays neatly on SI but is fading on DI like it's on the margin of DI coverage.
Now it's also likely that software use SI informations to display the widest part of DI beam, aiming at wider coverage, since in therory one should not be able to achieve such a wide beam using high frequency (the higher the frequency the narrower the beam, that's a physical law).
The way i see it, DI will cover accurately the underboat gap between both SI beams, but there has to be some overlaping (either naturally or by software calculation).
Here is a thread from other forum, where the beam coverage issue is well explained, especially in the 8th message from OThessalonikios. Espacially this sentence is important for a good understanding of how beams work : "In the flashlight analogy, it would be like telling someone "you can't see that object because it's at a place in the light 1/2 the intensity of the center", even though you can CLEARLY see it."