with t1 (id, date1, fld2, fld3) as
(
              select 111, to_date('10.10.2020', 'DD.MM.YYYY'), 10, 'вчера' from dual
    union all select 222, to_date('11.10.2020', 'DD.MM.YYYY'), 15, 'сегодня' from dual
    union all select 333, to_date('12.10.2020', 'DD.MM.YYYY'), 30, 'завтра' from dual
)
, t (rn, id, date1, fld2, fld3) as
(
    select row_number() over (order by date1) rn, id, date1, fld2, fld3 from t1
)
, rows_for_group (n) as
(
    select level-1 from dual connect by level < 5
)
select *
  from t, rows_for_group
 model dimension by (rn rnk, n) 
       measures (date1 dt,
                 cast (null as varchar2(100)) name,
                 fld2, fld3,
                 cast(null as number) f2,
                 cast(null as varchar2(20)) f3
                )
       rules AUTOMATIC ORDER
       (
         name[any, 0] = to_char(dt[cv(rnk), 0], 'DD.MM.YYYY'),
         name[any, 1] = 'Доп строка 1',
         name[any, 2] = 'Доп строка 2',
         name[any, 3] = 'Доп строка 3',
         f2[any, 0] order by rnk, n = coalesce(f2[cv(rnk)-1, 3], fld2[cv(rnk), 0]),
         f2[any, 1] order by rnk, n = 
             case when f2[cv(rnk), 0] + 3 < fld2[cv(rnk), 0]
                  then f2[cv(rnk), 0] + 3 else fld2[cv(rnk), 0] end,
         f2[any, 2] order by rnk, n = 
             case when f2[cv(rnk), 1] + 3 < fld2[cv(rnk), 1]
                  then f2[cv(rnk), 1] + 3 else fld2[cv(rnk), 1] end,
         f2[any, 3] order by rnk, n = 
             case when f2[cv(rnk), 2] + 5 < fld2[cv(rnk), 2]
                  then f2[cv(rnk), 2] + 5 else fld2[cv(rnk), 2] end,
         f3[any, 0] = '-----',
         f3[any, 1] = fld3[cv(rnk)-1, 0],
         f3[any, 2] = fld3[cv(rnk), 0],
         f3[any, 3] = fld3[cv(rnk)+1, 0],
         name[null, null] = 'Итого',
         f2[null, null] = sum(f2)[for rnk in (select rn from t), 
                                  for n in (select n from rows_for_group)]
       )
       order by rnk, n