While I'd never estimate mIRCs users to be in the upper millions, you shouldn't take the netsplit.de statistics for the "whole of IRC" (In a related thread, I pointed to the constant diversification of the network landscape and to the lack of interest or even open objection of many mid-size networks to be indexed).
Anyway, for a moment I'd like to put aside argumentats based on quantity only. If mIRC isn't the foremost app for "chat" any more, it shall remain the foremost app for IRC-based chat.

Digital world saw and will see the rise and fall of protocols, and of applications/implementations of these protocols. I don't suppose mIRC to have a tremendous peak again, or to become more popular than the zeitgeisty social networking thingies. But Imho IRC as protocol, and mIRC - as it's foremost client - is neither dead nor "quickly dying". It's alive and will continue to live, for a decent while at least. smile
Some distant day, IRC may be dead. But until that day I'd like mIRC to remain the superb IRC client it is, without attempts to turn it into something it never wanted to be. If some day IRC will be marginalized, mIRC will have become a niche tool - and I'd like mIRC still being mIRC that day.
Don't get me wrong, I wouldn't mind if new protocols were added who complement it's IRC-based capabilities. Yet mIRC should rely on it's own strenghs in the first place. For example DCC had always been subordinate to IRC in mIRC, in both the ammount of work/time Khaled put into developing features related to it and in the popularity of the protocol and features amongst mIRCs users.

Whatever new feature, it cannot compensate for a possibly diminishing interrest of users in IRC as a protocol, as a whole. Ergo mIRC shouldn't try too hard, and it should not lapse into imitations. All asymmetrical dispositions of resources for both advertising and developement put aside. Making mIRC at last a full-unicode app will for example open mIRC to many new languages and users.
And, so far, IRC/mIRC survived many "fashions" and proclaimed "revolutions", outlasted some of those "en vogue" apps and "social network" approaches.

As a final analogy (Yes, another analogy! :D): the success of TV over radio is undisputed. From some valid point of view, many users "migrated" from the one media to the other. And what should radio "have done about it" - it neither caused this shift of users nor missed it an existing chance to prevent it, e.g. by not being open minded for changes within the possibilities of the media. Many contemporaries declared radio to be dead at the end of the decade. But radio didn't die, and TV did not succeed radio as a media. The radio media is still alive, with it's own users, uses and characteristics. And I'm glad it is.