There are a few points that need to be pointed out-
Loco's and DVT take up platform length, so to obtain the same capacity from a train which is loco hauled all the main stations need to be lengthened to accommodate the extra length. Compare this to IEP needing works to be cleared on tight bends, much of which can be done or has been done when signals are replaced and gauge clearance works for freight is undertaken. As such to build and run a loco hauled train with the same capacity as an IEP is likely to cost more.
Cost comparisons per vehicle of a Class 390 and an IEP per vehicle are not comparing like with like. An 11 coach 390 has a capacity of 589 seats, whilst a 9 coach IEP has a capacity of 627. This then means that less IEP coaches are needed compared to 390 coaches (about 1/3), so assuming that each coach cost double the cost per seat is not, it is closer to 1.55 times the cost. If you work it backwards for the cost per seat to be the same the IEP coaches can cost 1.3 times that of the 390 coachs.
The length of an 11 coach 390 is 253m whilst a 9 coach IEP is 234m, meaning that by lengthening each station by just 7m more than would be needed for an 11 coach 390 it would enable a 10 coach IEP to call there. Which could mean that in the future the electric 9 coach IEP's could be lengthened to 10 coaches at a fairly low infrastructure cost (similar to the 11 coach lengthening of the 390 which has just happened). This would further increase capacity of the IEP's over and above the 390's, with a very minimal infrastructure cost, whilst to just to get up to 12 coaches 390s would cost a lot and would only put it a little ahead of the 9 coach IEP in terms of capacity.
The problem with buying diesel loco's is that it is likely that electrification will continue to spread, meaning a smaller and smaller network that they are suitable to run on, meaning that some could be redundant before they reach their 30th birthday, as well as needing to keep buying electric locos to cover each new part of the network that gets wired up.
Therefore bi-modal has it's uses when there is a program of extending the electrification of the network. For instance, what happens when a length of line gets electrified from a major station where a loco swap occurs because the local services justify electrifying the line? Do the IC trains then have to start stopping at some random station or do you run diesel trains under the wires for no reason? It also means that only long distances of electrification will benefit the IC services, rather than being able to wire up lengths as there is time/money/resources.
I can see that there are a lot of risks with IEP, however I think that they are often overplayed by those that don't like/want them for whatever reason. Likewise there would be a lot of risks (and probably a fair amount of objection) with using a train that is already being used on the UK rail network.