1.Creating a way to transfer funds into bitcoin system that would compete with credit card or PayPal in terms of convenience, e.g. you just need to sign up for credit card to start using it for payments.
Agreed, the ways to get money into the system are severely limited at the moment. Especially with Gox and TH having suspended SEPA transfers. Different from the other points, there is a big profit opportunity here, so I'm sure it will be solved sooner than the rest

2.Safe and secure e-wallet or
lightweight bitcoin client for smartphone to use on the go, capable of making instant transfer to use in store or to buy something from vending machine, etc. Something like
green address or
MtGox QR voucherAgreed. Smartphone support is really important. You might want to look at the Bitcoin-android client (
https://bt.irlbtc.com/view/26684.80). It still has some issues to iron out but it is pretty nice (and works with QR codes).
In addition to QR codes, NFC support would also be really nice.
3.Privacy and anonymity is important for long and
short term survival of bitcoin. De-anonymizing process should be made not cost effective by implementing a method that would enforce use of one address per transaction and having sophisticated algorithm to select inputs and outputs for transaction to prevent consolidation. Enforce use of some form of proxy for bitcoin client to hide IP address.
I don't think anonymity is very high up in the concerns, at least for the devs of the mainline client. There is certainly a place for a stealth bitcoin client (someone is working on selling bootable USB sticks with Bitcoin+TOR preloaded), however. It is not needed for mainstream adoption. We have enough shady types as it stands

4.Development of flexible software rather then all in one "take it or leave it" client, and eco system that would take advantage from it.
Yes.
There are various initiatives like this under way. I think the google one BitcoinJ (also used by the bitcoin-android client) is the most promising at the moment, but I know of a few more in active development (a Python client, another C++ client, there are designs for a scalable C one, etc..).